U.S. patent number 9,053,519 [Application Number 13/371,911] was granted by the patent office on 2015-06-09 for system and method for analyzing gis data to improve operation and monitoring of water distribution networks.
This patent grant is currently assigned to TAKADU LTD.. The grantee listed for this patent is Amitai Armon, Lilach Bien, Chaim Linhart, Haggai Scolnicov. Invention is credited to Amitai Armon, Lilach Bien, Chaim Linhart, Haggai Scolnicov.
United States Patent |
9,053,519 |
Scolnicov , et al. |
June 9, 2015 |
System and method for analyzing GIS data to improve operation and
monitoring of water distribution networks
Abstract
A computerized method for modeling a utility network. The method
includes retrieving geographical information system (GIS) data,
asset management data, and sensor archive data of one or more
assets of the utility network. The method also includes generating
one or more mathematical elements from the one or more assets and
creating probable connections between the one or more mathematical
graph elements based on the GIS and asset data. A mathematical
graph is generated based on the probable connections, the
mathematical graph including one or more asset characteristics of
the one or more assets. Analysis is performed on the utility
network using the mathematical graph and the mathematical graph
data is stored for use by other systems within the utility
network.
Inventors: |
Scolnicov; Haggai (Tel Aviv,
IL), Armon; Amitai (Tel Aviv, IL), Linhart;
Chaim (Petach Tikva, IL), Bien; Lilach (Rehovot,
IL) |
Applicant: |
Name |
City |
State |
Country |
Type |
Scolnicov; Haggai
Armon; Amitai
Linhart; Chaim
Bien; Lilach |
Tel Aviv
Tel Aviv
Petach Tikva
Rehovot |
N/A
N/A
N/A
N/A |
IL
IL
IL
IL |
|
|
Assignee: |
TAKADU LTD. (Yehud,
IL)
|
Family
ID: |
48946361 |
Appl.
No.: |
13/371,911 |
Filed: |
February 13, 2012 |
Prior Publication Data
|
|
|
|
Document
Identifier |
Publication Date |
|
US 20130211797 A1 |
Aug 15, 2013 |
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G06Q
50/06 (20130101); G06Q 10/0639 (20130101) |
Current International
Class: |
G06F
17/10 (20060101); G06Q 10/06 (20120101); G06Q
50/06 (20120101) |
Field of
Search: |
;703/2 |
References Cited
[Referenced By]
U.S. Patent Documents
Foreign Patent Documents
|
|
|
|
|
|
|
102008048748 |
|
Sep 2008 |
|
DE |
|
0298479 |
|
Jul 1988 |
|
EP |
|
0521548 |
|
Jan 1993 |
|
EP |
|
0821817 |
|
Jun 1999 |
|
EP |
|
1047925 |
|
Mar 2004 |
|
EP |
|
0925491 |
|
May 2004 |
|
EP |
|
1324165 |
|
Nov 2007 |
|
EP |
|
1867970 |
|
Dec 2007 |
|
EP |
|
1133690 |
|
Nov 2008 |
|
EP |
|
2060896 |
|
May 2009 |
|
EP |
|
2478432 |
|
Mar 2011 |
|
GB |
|
1020080005694 |
|
Jan 2008 |
|
KR |
|
439028 |
|
Jun 2001 |
|
TW |
|
200713114 |
|
Apr 2007 |
|
TW |
|
0020832 |
|
Apr 2000 |
|
WO |
|
0194937 |
|
Dec 2001 |
|
WO |
|
0195227 |
|
Dec 2001 |
|
WO |
|
0245315 |
|
Jun 2002 |
|
WO |
|
02086670 |
|
Oct 2002 |
|
WO |
|
2004113863 |
|
Dec 2004 |
|
WO |
|
2005094493 |
|
Oct 2005 |
|
WO |
|
2007002838 |
|
Jan 2007 |
|
WO |
|
2008016697 |
|
Feb 2008 |
|
WO |
|
2008029681 |
|
Mar 2008 |
|
WO |
|
2008067442 |
|
Jun 2008 |
|
WO |
|
2008094551 |
|
Aug 2008 |
|
WO |
|
2009055342 |
|
Apr 2009 |
|
WO |
|
2010035281 |
|
Apr 2010 |
|
WO |
|
Other References
UM. Shamsi, "GIS Applications for Water, Wastewater, and Stormwater
Systems," 2005, Taylor & Francis, 48 pages. cited by examiner
.
"Thames Water Utilities," 2007, 259 pages. cited by examiner .
Alexandru Aldea et al., "Network modeling solution for water supply
using GIS concepts," 2008, Conferinta international: "Solutii
pentru sisteme de alimentaire cu apa si canalizare in localitati
pana la 10000 locuitori," pp. 202-208. cited by examiner .
"GIS Implementation for Water and Wastewater Treatment Facilities,"
2005, Water Environment Federation, pp. 81-107. cited by examiner
.
S. S. Skiena, "The Algorithm Design Manual," 2008, second edition,
Springer-Verlag, pp. 145-190, 475-522, 523-561. cited by examiner
.
Judith L. Gersting, "Mathematical structures for computer science,"
sixth edition, 2007, W.H. Freeman and Company, pp. 520-527. cited
by examiner .
"Water Infrastructure Monitoring," 2010, Wayback Machine entry for
http://www.takadu.com/files/Brochure/TaKaDuOnePager.pdf, two pages.
cited by examiner .
Samir Khuller et al., "Fault tolerant K-center problems," 2000,
Theoretical Computer Science, vol. 242, pp. 237-245. cited by
examiner .
Ines Slama et al. "Energy efficient scheme for large scale wireless
sensor networks with multiple sinks," 2008, IEEE Wireless
Communications and Networking Conference 2008, pp. 2367-2372. cited
by examiner .
"NetworkX GIS Shapefile," 2011,
http://networkx.lanl.gov/reference/readwrite.nx.sub.--shp.html, one
page. cited by examiner .
Bin Jiang et al., "Topological analysis of urban street networks,"
2004, Environment and Planning B: Planning and design, vol. 31, pp.
151-162. cited by examiner .
Alireza Yazdani et al., "Complex network analysis of water
distribution systems," Apr. 2011, arXiv, pp. 1-11. cited by
examiner .
Lina Perelman et al., "Topological clustering for water
distribution systems analysis," available online Feb. 15, 2011,
Environmental Modelling & Software, vol. 26, pp. 969-972. cited
by examiner .
Danilo Mandic, et al. "Data Fusion for Modern Engineering
Application: An Overview", ICANN 2005, LNCS 3697, pp. 715-721
(2005). cited by applicant .
Sholom M. Weiss and Nitin Indurkhya. Predictive Data Mining: A
Practical Guide. Morgan Kaufmann, 1997, p. 147. cited by applicant
.
ABB Group (2009) "Solutions for the Water Cycle Leakage
Management", ABB Water Industrial Sector Initiative, 26 slides
(pages); state retrieved from
http://www.abb.com/industries/db0003db004063/a0a4683761017bb2c12574b-
b002be3b5.aspx. cited by applicant .
Stephen Mounce, et al. (2006) "Burst Detection Using Hydraulic Data
From Water Distribution Systems With Artificial Neural Networks",
Urban Water Journal, pp. 12 pages, UK. cited by applicant .
Dalius Misiunas (2005) "Failure Monitoring and Asset Condition
Assessment in Water Supply Systems", Doctoral Dissertation in
Industrial Automation Department of Industrial Electrical
Engineering an Automation, Lund University, Sweden, 349 pages.
cited by applicant .
Centre for Water Systems: A Leading Center for Water Engineering
Research in the UK. vol. II, Issue 2, Autumn 2009
<http://centres.exeter.ac.uk/cws/downloads/doc.sub.--download/53-cws-n-
ewsletter-volume-2-issue-2>. cited by applicant .
Frenk Withoos, et al. (2004) "Leak Manager, Analysis and Management
of Water Leakage in Potable Water Distribution Networks Using
Advise IT Water Leakage Management", ABB Review, Innovative
Engineering, pp. 61-63; state retrieved from
http://www.abb.com/industries/db0003db004063/a0a4683761017bb2c12574bb002b-
e3b5.aspx. cited by applicant .
Ivan Stoianov, et al. (2007) Pipenet: A Wireless Sensor Network for
Pipeline Monitoring, In IPSN07: 264-273. ACM, 2007, USA, 10 pages.
cited by applicant .
Gabrys, B. and Bargiela, A. "Simulation of Water Distribution
Systems". In: The European Simulations Symposium (ESS'98), Oct.
26-28, 1998, Nottingham, UK, pp. 273-277. cited by applicant .
Mamlock, R. and Al-Jayyousi, O. "Fuzzy Sets Analysis for Leak
Detection in Infrastructure Systems: A Proposed Methodology". In:
Clean Technology Environmental Policy 6, pp. 26-31, 2003. cited by
applicant.
|
Primary Examiner: Craig; Dwin M
Assistant Examiner: Guill; Russ
Attorney, Agent or Firm: Ostrow, Esq.; Seth H. Meister
Seelig & Fein LLP
Claims
What is claimed is:
1. A computer-implemented method for modeling a utility network,
the method comprising: retrieving geographical information system
(GIS) data and asset management data of one or more assets of the
utility network, wherein the GIS data and asset management data do
not indicate connections between the one or more assets and wherein
the GIS data includes coordinate data associated with the one or
more assets; generating, via a processing device, one or more
mathematical graph elements from the one or more assets; creating,
via the processing device, probable connections between the one or
more mathematical graph elements based on the GIS and asset
management data, wherein creating probable connections comprises
snapping a plurality of junctions based on coordinate data;
generating, via the processing device, a mathematical graph based
on the probable connections, the mathematical graph including one
or more asset characteristics of the one or more assets; analyzing,
via the processing device, the determined junctions, wherein
analyzing comprises determining if an analyzed junction appears
between only two other junctions and merging the two other
junctions if the analyzed junction appears between only two other
junctions; identifying, via the processing device and by analyzing
the mathematical graph, one or more flow monitoring zones (FMZs) in
the utility network, wherein identifying one or more FMZs is based
upon analyzing junction locations; and storing the mathematical
graph data for use by one or more systems.
2. The method of claim 1, wherein creating probable connections
includes creating probable connections based on an analysis of an
overlay of a plurality of GIS layers.
3. The method of claim 2, wherein the GIS layers include an asset
layer, a pipe layer, and a zone layer.
4. The method of claim 1, wherein the one or more mathematical
graph elements includes at least one of nodes, edges, and
polygons.
5. The method of claim 1, wherein the mathematical graph includes
at least one of a directed graph, undirected graph, mixed graph,
multi-graph, simple graph, weighted graph, and Cartesian graph.
6. The method of claim 1, wherein generating the mathematical graph
includes at least one of a graphical, geometrical, numerical,
differential, functional, and algebraic analysis.
7. The method of claim 1 further comprising determining one or more
consumption equations from analysis of the mathematical graph.
8. The method of claim 1, wherein identifying FMZ characteristics
includes identifying split FMZs, almost split FMZs, single point of
failure, and neighboring FMZs.
9. The method of claim 1 further comprising generating warnings and
suggested maintenance recommendations.
10. The method of claim 9, wherein generating warnings and
suggested maintenance recommendations comprises presenting one or
more physical locations on the mathematical graph.
11. The method of claim 1 further comprising locating optimal meter
locations.
12. The method of claim 11, wherein locating optimal meter
locations comprises presenting one or more physical locations on
the mathematical graph.
13. The method of claim 11, wherein locating optimal meter
locations comprises performing balanced cuts on the mathematical
graph to partition a flow monitoring zone (FMZ) into sub-areas.
14. The method of claim 11, wherein locating optimal meter
locations comprises identifying k-center nodes on the mathematical
graph.
15. The method of claim 1, wherein generating a mathematical graph
comprises arranging the one or more mathematical graph elements on
the mathematical graph based on the created connections.
16. A system for modeling a utility network, the system comprising:
a memory device having executable instructions stored therein; and
a processing device, in response to the executable instructions,
operative to: retrieve geographical information system (GIS) data
and asset management data of one or more assets of the utility
network, wherein the GIS data and asset management data do not
indicate connections between the one or more assets and wherein the
GIS data includes coordinate data associated with the one or more
assets; generate, via the processing device, one or more
mathematical graph elements from the one or more assets; create,
via the processing device, probable connections between the one or
more mathematical graph elements based on the GIS and asset
management data, wherein creating probable connections comprises
snapping a plurality of junctions based on coordinate data;
generate, via the processing device, a mathematical graph based on
the probable connections, the mathematical graph including one or
more asset characteristics of the one or more assets; analyze the
determined junctions, wherein analyzing comprises determining if an
analyzed junction appears between only two other junctions and
merging the two other junctions if the analyzed junction appears
between only two other junctions; identify one or more
characteristics of one or more flow monitoring zones (FMZs) in the
utility network, wherein identifying one or more FMZs is based upon
analyzing junction locations; and store the mathematical graph data
for use by one or more systems.
17. The system of claim 16, wherein GIS data includes one or more
GIS layers.
18. The system of claim 17, wherein the GIS layers include an asset
layer, a pipe layer, and a zone layer.
19. The system of claim 16, wherein the one or more mathematical
graph elements includes at least one of nodes, edges, and
polygons.
20. The system of claim 16, wherein the processing device is
further operative to determine one or more consumption equations
from analysis of the mathematical graph.
21. The system of claim 16, wherein the processing device is
further operative to identify split FMZs, almost split FMZs, single
point of failure, and neighboring FMZs.
22. Non-transitory computer readable media comprising program code
that when executed by a programmable processor causes execution of
a method for modeling a utility network, the computer readable
media comprising: computer program code for retrieving geographical
information system (GIS) data and asset management data of one or
more assets of the utility network, wherein the GIS data and asset
management data do not indicate connections between the one or more
assets and wherein the GIS data includes coordinate data associated
with the one or more assets; computer program code for generating,
via the processor, one or more mathematical graph elements from the
one or more assets; computer program code for creating, via the
processor, probable connections between the one or more
mathematical graph elements based on the GIS and asset management
data, wherein creating probable connections comprises snapping a
plurality of junctions based on coordinate data; computer program
code for generating, via the processor, a mathematical graph based
on the probable connections, the mathematical graph including one
or more asset characteristics of the one or more assets; computer
program code for analyzing, via the processor, the determined
junctions, wherein analyzing comprises determining if an analyzed
junction appears between only two other junctions and merging the
two other junctions if the analyzed junction appears between only
two other junctions; computer program code for identifying, by
analyzing the mathematical graph, one or more flow monitoring zones
(FMZs) in the utility network, wherein identifying one or more FMZs
is based upon analyzing junction locations; and computer program
code for storing the mathematical graph data for use by one or more
systems.
23. The computer readable media of claim 22, wherein the computer
program code for identifying FMZ characteristics includes computer
program code for identifying split FMZs, almost split FMZs, single
point of failure, and neighboring FMZs.
24. The computer readable media of claim 22 further comprising
computer program code for locating optimal meter locations.
25. The computer readable media of claim 24, wherein the computer
program code for locating optimal meter locations further comprises
computer program code for performing balanced cuts on the
mathematical graph to partition a flow monitoring zone (FMZ) into
sub-areas.
26. The computer readable media of claim 24, wherein the computer
program code for locating optimal meter locations further comprises
computer program code for identifying k-center nodes on the
mathematical graph.
27. The computer readable media of claim 22, wherein the computer
program code for generating the mathematical graph further
comprises program code for arranging the one or more mathematical
graph elements on the mathematical graph based on the created
connections.
Description
CROSS REFERENCE TO RELATED APPLICATIONS
The present application is related to: U.S. patent application Ser.
No. 12/717,944, entitled "SYSTEM AND METHOD FOR MONITORING
RESOURCES IN A WATER UTILITY NETWORK," filed on Mar. 5, 2010, now
issued as U.S. Pat. No. 7,920,983; U.S. patent application Ser. No.
13/008,819, entitled "SYSTEM AND METHOD FOR IDENTIFYING LIKELY
GEOGRAPHICAL LOCATIONS OF ANOMALIES IN A WATER UTILITY NETWORK,"
filed on Jan. 18, 2011; and U.S. patent application Ser. No.
13/313,261, entitled "SYSTEM AND METHOD FOR IDENTIFYING RELATED
EVENTS IN A RESOURCE NETWORK MONITORING SYSTEM," filed on Dec. 7,
2011; the disclosures of which are hereby incorporated herein by
reference in their entirety.
COPYRIGHT NOTICE
A portion of the disclosure of this patent document contains
material, which is subject to copyright protection. The copyright
owner has no objection to the facsimile reproduction by anyone of
the patent document or the patent disclosure, as it appears in the
Patent and Trademark Office patent files or records, but otherwise
reserves all copyright rights whatsoever.
FIELD OF THE INVENTION
The invention described herein generally relates the field of
monitoring and operating resource distribution networks such as
utility water distribution networks and, in particular, to
analyzing and using GIS data to improve resource network operation
and monitoring.
BACKGROUND OF THE INVENTION
The United Nations notes that water use has been growing at more
than twice the rate of population increase in the last century, and
an increasing number of regions are chronically short of water. By
2025 two-thirds of the world's population could be under water
stress conditions as a result of population growth and other
factors. Water, especially potable water, is essential for all
socio-economic developments and for maintaining a healthy
population. As populations increase across the globe they call for
an increased allocation of clean water for use, resulting in
increased water scarcity.
A significant amount of water may be conserved merely by addressing
the loss of water or degradation in water quality in systems caused
by leaks or other adverse effects. Thus, one method of addressing
water scarcity and conserve resources is to improve the operation
and monitoring of the utility networks used to deliver water, such
as by faster and more accurate detection of leaks and other events
occurring in such networks. Several systems currently exist to
facilitate improved network monitoring in water utility networks.
For example, commonly owned U.S. Pat. No. 7,920,983, entitled
"SYSTEM AND METHOD FOR MONITORING RESOURCES IN A WATER UTILITY
NETWORK" which is herein incorporated by reference in its entirety,
describes sophisticated systems and methods for detecting anomalies
in water utility networks using statistical and analytical
techniques, some of which are in use by the assignee of the present
invention, TaKaDu Ltd. Other systems available from other
companies, such as those available from ABB Group or IBM Corp.,
also provide some improvements to anomaly detection in water
utility networks.
One way in which the operation and monitoring of water utility
networks may be further improved is by making better use of data
from Geographical Information Systems ("GIS"s or "GIS systems") and
asset management systems. As known to those of skill in the art, a
GIS integrates, stores, and displays geographic information about a
network or system laid out in a physical environment. Applications
using GIS allow users to create interactive queries, review spatial
information, edit data and maps, and present the results of these
operations in a graphical user interface. Further description and
details of GIS systems may be found in "Getting Started with
Geographic Information Systems," Second Edition by Keith C. Clarke,
which is hereby incorporated by reference in its entirety. As is
also known to those of skill in the art, asset management systems
store information about physical components of a network or system
such as a water utility network, such as pipes or joints, and are
used in the operation and management of networks such as in fixing
network components or in ordering new or replacement components.
Commercially available asset management systems used to manage
water utility network assets are available from a variety of
entities, as known to those of skill in the art.
Water utilities (or other network operators) currently use data
from GIS and asset management systems to display the physical
layout of their distribution networks and identify characteristics
of individual assets in the networks. Often times, GIS data is
merely displayed and not used for automated analysis and functions.
The usage of GIS data has great potential for planning network
improvements, continuous on-line monitoring, and improving the
understanding of a network's current configuration. However,
effective use of GIS and asset data is not being made in current
systems that monitor utility networks.
As such, there exists a need for improved systems and methods for
automated modeling and analysis of networks and network components
using GIS and asset management systems. The present invention
provides for improved analysis and usage of GIS data in utility
network monitoring systems.
SUMMARY OF THE INVENTION
Methods and systems are provided for improved modeling, monitoring
and operation of a utility network. One such method includes
retrieving geographical information system (GIS) data, asset
management data, and sensor archive data of one or more assets of
the utility network. The method also includes generating one or
more mathematical graph elements from the one or more assets and
creating probable connections between the one or more mathematical
graph elements based on the GIS and asset management data. A
mathematical graph is generated based on the probable connections
to obtain more complete GIS data. The mathematical graph includes
one or more asset characteristics of the one or more assets.
Analysis is performed on the utility network using the mathematical
graph and the mathematical graph data of the utility network is
stored for use by other systems within the utility network.
In some embodiments, the GIS data includes one or more GIS layers.
GIS layers may further include an asset layer, a pipe layer, and a
zone layer. Mathematical graph elements include at least one of
nodes, edges, and polygons. The generated mathematical graph
includes at least one of a directed graph, undirected graph, mixed
graph, multi-graph, simple graph, weighted graph, and Cartesian
graph. Analysis performed on the mathematical graph includes at
least one of a graphical, geometrical, numerical, differential,
functional, and algebraic analysis.
In certain embodiments, performing analysis on the utility network
using the mathematical graph includes determining one or more
consumption equations from the analysis of the mathematical graph.
Performing analysis on the utility network using the mathematical
graph may also include identifying flow monitoring zone (FMZ)
characteristics of the utility network. Identifying FMZ
characteristics include identifying split FMZs, almost split FMZs,
single point of failure, and neighboring FMZs. Performing analysis
on the utility network using the mathematical graph may also
include generating warnings and suggested maintenance
recommendations. Generating warnings and suggested maintenance
recommendations includes presenting one or more physical locations
on the mathematical graph.
Performing analysis on the utility network using the mathematical
graph may also include locating optimal meter locations. Locating
optimal meter locations includes presenting one or more physical
locations on the mathematical graph. In one embodiment, locating
optimal meter locations includes performing balanced cuts on the
mathematical graph to partition a FMZ into sub-areas. In another
embodiment, locating optimal meter locations includes identifying
k-center nodes on the mathematical graph. Generating the
mathematical graph structure including the plurality of
mathematical graph elements may include arranging the plurality of
mathematical graph elements on the mathematical graph based on the
created connections.
In another aspect, the present invention provides for prioritizing
areas for maintenance in a utility network. The method includes
receiving a training set of historical leak data of the utility
network. The method also includes receiving one or more feature
values for the historical leak or failure data, the one or more
feature values including factors that influence leakage rates and
generating a model from the training set and the one or more
feature values. The model is provided with one or more current leak
locations within the utility network and associated feature values,
and predicted leakage or failure rates are obtained for the one or
more current leak locations from the model.
According to one embodiment of this aspect of the present
invention, the predicted leakage rates can be utilized to determine
field work priority wherein higher predicted leakage rates are
given higher priority. Retrieving one or more feature values
associated with the network areas comprises analyzing GIS data,
asset data, and historical repair data. The one or more feature
values may include age of hardware, pipe fittings, and valves.
Feature values may also include network complexity. The network
complexity may include the amount of pipes, fittings, valves, and
service connections. The one or more feature values may further
include geographical properties that contribute to more leaks being
hidden, pressure based on topography and pressure meter readings,
rate of leaks estimated from historical repair files and
computational analysis of historical data from flow and pressure
meters, leak frequency, loss rate, hidden time, and total costs.
Generating a model from the one or more feature values of the
training set may also include using a least one of linear
regression, non-linear regression, decision trees, neural networks,
k-nearest neighbor, and support vector machines.
BRIEF DESCRIPTION OF THE DRAWINGS
The invention is illustrated in the figures of the accompanying
drawings which are meant to be exemplary and not limiting, in which
like references are intended to refer to like or corresponding
parts, and in which:
FIG. 1 presents a block diagram of a computing system according to
an embodiment of the present invention.
FIG. 2 presents a flow diagram of a method for modeling a water
utility network according to an embodiment of the present
invention.
FIGS. 3 and 4 present flow diagrams of a method for pre-processing
asset management and GIS data according to an embodiment of the
present invention.
FIGS. 5 and 6 present network graphs for characterizing water
sub-networks according to certain embodiments of the present
invention.
FIG. 7 presents a flow diagram of a method for predicting leakage
rates for prioritizing areas of maintenance according to an
embodiment of the present invention.
DETAILED DESCRIPTION OF THE INVENTION
In the following description of the embodiments of the invention,
reference is made to the accompanying drawings that form a part
hereof, and in which is shown by way of illustration, exemplary
embodiments in which the invention may be practiced. It is to be
understood that other embodiments may be utilized and structural
changes may be made without departing from the scope of the present
invention.
FIG. 1 presents a block diagram illustrating one embodiment of a
system for analyzing and monitoring assets and their geographical
layout in a water distribution system. As shown in FIG. 1, the
system includes a structural analysis system 102 operable to
retrieve data from several sources including asset management
database 103, geographic information system (GIS) 106, and sensor
archive database 107, to analyze the retrieved data in accordance
with processes described herein, and to output the resulting data
to enriched GIS database 101. The enriched GIS data 101 can then be
used in a number of ways including for improved analysis of
anomalies and events by a water network analysis engine 104, and by
other systems, all as further described herein.
Asset management database 103 stores information on assets in the
water distribution system, including pipes, valves, meters and
other components that make up the distribution network. Asset
management database 103 may be part of an asset management system
such as the Maximo.TM. system available from IBM Corp. Asset
management data may be any information associated with network
assets relating to infrastructure and inventory of a network. For
example, asset management data may include information such as age,
size, shape, length, diameter, material, and other characteristics,
concerning pipes, line segments, valves, and meters installed in
the network. Asset management data may further include data which
indicates an anomaly, such as consumer reports of service failures,
or sightings of a visible burst, administrator or operational
divisions of the network, information concerning water utility
network operations, such as routine or planned water utility
network operations, opening and closing of valves that affect water
flow, pump operations, acoustic surveys, repairs or improvements
made to any part of the water utility network, dates and times of
the repairs or improvements, locations of the repairs or
improvements, routine maintenance made to the network, and access
control information indicating when and where on the network
technical personnel may be active. The data in asset management
database 103 is typically used to help manage the water utility
network.
GIS 106 provides GIS data which describes the structure and layout
of the water utility network and the positioning of the meters
across it. GIS data may include: descriptions of the water pipes
such as the diameter, length, installation date and manufacturing
materials; meter types, meter locations, and meter ages; partitions
of the network into pressure zones and/or supply zones; a city or
area map; and additional evolving data recognized by one of skill
in the art. GIS 106 may also record features of a water
distribution system including pipes, valves, pumps, treatment
plant, reservoirs, storage tanks, etc., as well as customer and
pipe locations and past usage, elevations, etc. GIS may further
store data regarding portions of the network such as District Meter
Areas ("DMA"s) or Flow Monitoring Zones ("FMZ"s). By way of
background, utilities companies often divide their utility networks
into several DMAs for identifying problematic areas that may
contain leaks. A DMA is usually a sub-network that is partitioned
by geographical area (such as a polygon on a map where the
customers served by it are located) or inlet/outlet points. All
inlet/outlet points are monitored by flow meters, allowing the
total flow into that sub-network to be deduced. Sub-networks with
such "total flow monitoring" may also be created as a by-product of
network planning or maintenance and this description refers to all
such sub-networks, inadvertent or deliberate, as Flow Monitoring
Zones, or FMZs. FMZs may result from a division or an inadvertent
split of DMAs that may have in turn resulted from valve closures or
the maintenance or installation of new pipes. FMZs divide a utility
network into manageable sections in order to make it easier for
engineers to determine the occurrence and location of bursts and
other faults and to repair them.
In some systems, GIS data from GIS 106 represent pipes as layers or
collections of line segments or broken lines, represent sensors as
points, and represent larger objects or areas such as a tank, DMA,
or water treatment plant as polygons. Assets' locations may be
represented using x-y coordinates (or x-y-z, including elevation),
e.g. one set for a point asset, two for the ends of a line segment,
or several for the vertices of a polygon. Hierarchical
relationships between GIS objects may also be recorded, for example
storing for each DMA a table of identifiers of the assets which
form part of that DMA, and recording for each of those assets that
it belongs to that DMA.
At least a portion of the GIS data may be retrieved or derived from
asset management data 103 and GIS 106. Data obtained from GIS 106,
sensor archive data 107, and asset management data 103 may be
non-live/non-real-time data such as data that has been archived or
offline data that includes data of typical operating conditions.
Non-live data can be static snapshots of the water utility network
that may be updated regularly or periodically. It is also noted
that this data may be evolutionary data including updates
consistent with the evolution of the underlying resource system
itself, for example, when new water pipes, connections, meters,
etc., are installed or otherwise modified in the system.
Furthermore, this data may include updates when the underlying
resource system is sampled or measured, for example when existing
pipes are inspected for material fatigue or internal constriction
by accumulated solid deposits. Any other characteristics of the
geography and engineering of the water distribution system may also
be utilized, as well as any other data relied on by one skilled in
the art.
Structural analysis system 102 is a computer system capable of
capturing, storing, analyzing, managing and presenting data with
reference to geographic location data provided by GIS 106. The
structural analysis system 102 links or integrates information that
is difficult to associate through any other means, and analyses
this information to improve the quality of GIS data and gain new
insight into network structure, not readily available through any
other means. For example, structural analysis system 102 is
programmed to merge GIS data from GIS 106 with asset management
data 103 and sensor archive data 107 to create the enriched GIS
data stored in database 101. Structural analysis system 102 can use
combinations of mapped variables from other data sources to build
and analyze new variables. The processes employed in structural
analysis system 102 to analyze and merge this data are set forth in
greater detail below. Structural analysis system 102 consists of
various software modules and databases residing on computer
hardware and performing the functions described further below.
Although illustrated as a single system, in various embodiments the
illustrated system may be integrated and/or distributed across
multiple hardware devices or processors and may be distributed
logically, physically or geographically. Structural analysis system
102 may be any suitable physical processing device performing
processing operations as described herein, in response to
executable instructions.
As shown in FIG. 1, the enriched GIS data 101 generated by
structural analysis system 102 is used by other systems including a
water network analysis engine 104 and the original GIS 106. In
addition, the output from the structural analysis system 102 is
provided to various output and reporting modules including a
geographical/structural user interface 105, and a reporting system
108. The operation of these modules and the use they make of the
output of the structural analysis system 102 is described further
below.
The water network analysis engine 104 analyzes data received from
different meters, sensors, reading devices, or other data sources
pertaining to a distribution network, detects anomalous data and
classifies some detected anomalies as events. One of skill in the
art will appreciate that unless the specific context explicitly
indicates otherwise, as used herein the terms "meter," "sensor,"
and "logger" generally refer to the same class of network devices
and encompass any meter, sensor, gauge, or other device capable of
measuring parameters or values or a stimulus, especially stimuli
relating to a water distribution network. The water analysis engine
104 provides automated analysis of the water distribution network
based on the received data, and provides real-time alerts, off-line
data reports and network planning recommendations to users who can
then take action. The types of event detected by the water network
analysis engine 104 include leaks, bursts, water consumption
changes, faulty meters, meter calibration problems, water quality
changes, malfunctions in network devices, asset utilization,
sub-network characteristics, and other events known to those
skilled in the art. Analysis engine 104 may be any network analysis
system suitable for use with a water treatment plant, which may be
any suitable resource distribution network, such as a municipal,
rural, or wholesaler water utility network, liquid distribution
network in a factory or other large building, or naval vessel, or
any suitable resource collection network such as a sewer system. An
example of a water network analysis engine includes the one
described in previously mentioned and commonly owned U.S. Pat. No.
7,920,983 and offered from the assignee TaKaDu Ltd., though
analysis engines available from other entities may also be used
within the context of the present invention.
In one embodiment, structural analysis system 102 stores,
manipulates, and reports distribution regions in the form of
predefined network parts such as DMAs or pressure zones. This
information may be presented as polygons on a geographical map,
ranges of addresses, or sets of marked or named network assets such
as lengths of pipe. Reports may be generated at reporting system
108 from a combination and processing of the aforementioned data
from GIS 106, sensor archive data 107, and asset management data
103. Data from structural analysis system 102 may also be retrieved
by one or more interface systems on geographical/structural user
interface 105. For example, the data may be retrieved by a trouble
ticket interface system to inform maintenance personnel of leaks or
other events. An example of trouble ticketing software is
Numara.RTM.'s Track-It!.RTM.. As another example, the event data
may be sent to a workflow interface system. One example of a
workflow system interface is Handysoft's Bizflow.RTM.. Event data,
for event reporting to users, is well categorized and can be
adapted for use by any industry standard interface. In addition to
GIS information, the geographical/structural user interface 105 may
receive automated recommendations and analysis of GIS data, asset
management, map, and various other data input streams used by the
structural analysis system 102, as additional layers. Information
received by geographical/structural user interface 105 is stored
and, in some embodiments, may be fed back into structural analysis
system 102 or used to supplement enriched GIS database 101.
Different types of interface systems may be used to provide
information on events to users or external systems in different
ways. The geographical/structural user interface 105 may be
accessed by various computerized devices, such as desktop computers
and laptops, cell phones, blackberry devices, smart phones, pagers
and other mobile devices programmed to receive pages, trouble
tickets and other types of alerts. The user interface 105 may be
accessed by computerized devices requesting it from servers
connected over any suitable network, such as the Internet, or may
be pushed out to such devices for viewing by users or input into
other systems such as trouble tickets systems. Outputs from water
network analysis engine 104 and/or structural analysis system 102
may be stored in the enriched GIS database 101, in an electronic
log file, and/or printed to paper. Previously stored data may be
accessed from storage to provide continuity in the reporting of
events, for example to update that a previously detected event is
still ongoing, rather than detecting it as an additional, separate
event. Previously stored data may also be used to generate training
sets for predicting future leakage rates or costs, and testing.
In accordance with aspects of the present invention, data provided
by structural analysis system 102 is displayable as mathematical
graphs on geographical/structural user interface 105. Mathematical
graphs include linear networks of objects that can be used to
represent interconnected features and to perform spatial analysis
on them. A mathematical graph may be composed of edges connected at
nodes or points, similar to graphs in mathematics and computer
science, and may include single dimension non-planar graphs with
the edge and node elements connected by a topology with additional
features and characteristics of the nodes and edges displayed on
the graph. According to the embodiments of the present invention,
pipes may be represented as edges connected to features including
other pipes, valves, meters, hydrants, and other network assets. As
with other types of graphs, a network graph can have
characteristics such as numerical weights and flows assigned to its
edges, which can be used to represent the various interconnected
features more accurately. Mathematical graphs may similarly be used
to model road networks and other resource utility networks, such as
electric or gas utility networks.
GIS data retrieved from GIS 106 is transformed into mathematical
graph structures by various pre-processing of asset information by
structural analysis system 102. Transformation and pre-processing
of asset and GIS data are described in further detail below with
respect to FIG. 2. By transforming the GIS data into a mathematical
graph structure, characterizing the connectivity and parameters of
the various network assets, and by applying standard graph-based
algorithms to the data in GIS layers such as pipes, assets, and
zones or DMAs, a network trace of each FMZ or DMA may be obtained.
The trace may include all the pipes, valves and other network
assets that illustrate how a real or actual DMA may be laid out as
a result of the pre-processing. Using the traces and auxiliary
information from the GIS, interesting insights about the network
may be extracted, some of which may also improve monitoring
capabilities of water network analysis engine 104. As described in
further detail below, analysis of GIS data may be used to provide
FMZ consumption equations or formulas, more accurately predict
leakage rates and test for leaks, assist with network planning and
installing pipes, make better recommendations for performing
network maintenance, configuring networks, and installing
additional meters to improve data for network analysis.
FIG. 2 presents a flow diagram of a method for modeling a water
utility network according to an embodiment of the present
invention. In step 201, GIS data, asset management data, and sensor
archive data of one or more assets are retrieved from a GIS
database, asset management system and/or an archive of static
snap-shots of GIS layers. The GIS layers may include asset layers,
pipe layers, and regional layers such as subdivision into DMAs,
pressure zones, and supply zones. Retrieval of GIS and asset data
may include retrieval of points, locations, and borders
representative of the assets and their physical locations or
approximate physical locations in a region. One or more
mathematical graph elements representing the one or more assets are
generated from the retrieved data, step 203. Generating the one or
more mathematical graph elements involves, in some embodiments,
generating abstract representations of the retrieved data such as
generating a list of nodes corresponding to certain types of assets
such as sensors, valves or connections, generating lists of edges
representing pipes, and generating polygons to illustrate larger
objects such as water tanks, a treatment plant or DMAs.
Probable connections are created between the one or more
mathematical graph elements based on the GIS and asset management
data, step 205. As previously mentioned, information available from
GIS and asset management data generally do not provide connections
between assets. Therefore, pre-processing of asset and GIS data is
performed. As described further with reference to FIGS. 3-5,
pre-processing includes transforming assets and other items from
the GIS data into a list of nodes and creating probable connections
by analyzing an overlay of various GIS layers. A table is also
created storing the probable connections established by the
pre-processing including pipe/asset ID, a start point, and end
point, diameter of the pipe, age, and other asset management data
previously mentioned. From the table, associations or connections
may be made between where different assets connect. For example,
pipes may be connected to each other, along with any meters,
valves, hydrants, etc. All of these assets may be grouped or fitted
into a polygon representing a given DMA indicating that the assets
are all within the given DMA, and this grouping of assets and their
associated connections is stored in a DMA table, with long pipes
extending through but not connected to any pipes of a DMA also
reflected in or determined contextually from the data included in
the DMA table. This process is further described with reference to
FIG. 4.
In a next step 207, a mathematical graph is generated based on the
created connections including the characteristics of the one or
more assets. A mathematical graph of the elements may be a directed
graph, undirected graph, mixed graph, multi-graph, simple graph,
weighted graph, or Cartesian graph. Because the mathematical graph
elements are arranged in the mathematical graph based on their
connectivity and other associations with one another, the graph
resembles a map or network of connected assets.
Step 209 includes storing the mathematical graph data for analysis
and for use by other systems. These other systems may include a
geographical/structural user interface and a reporting system. The
mathematical graph may also be provided to a network analysis
engine, a database for storage, or used to supplement a GIS system.
The connections created between the one or more mathematical graph
elements enables mathematical analysis on the mathematical graph.
Mathematical analysis performed on the mathematical graph may
include graphical, geometrical, numerical, differential,
functional, and algebraic analysis. The analysis performed on the
mathematical graph may be used to identify DMA characteristics,
generate warnings, suggest maintenance recommendations, and locate
optimal meter locations, which are described in further detail
below.
To generate the best mathematical graph from the data available, a
best-effort type of matching is made to determine the connectivity
of nodes and edges. Matching may include looking for edges with
ends that are very close or the closest to each other and puzzling
or connecting the ends together to identify the most probable
connections. The determination of closeness may be bounded by a
threshold, e.g., less than x number of feet apart. The threshold
may be determined by a user of the GIS analysis system or set to a
predefined value in the structural analysis system based on certain
parameters and region of the assets. In one embodiment, a user may
be prompted to verify assets with questionable connections. The
connectivity and associations between the features of the map may
be extracted and stored in an index with the edges and nodes,
numbers, or fractions, etc.
One embodiment of this method for pre-processing asset management
and GIS data according to the present invention is illustrated in
FIG. 3. A first phase of the pre-processing, sometimes referred to
herein as snapping, starts at step 301, in which coordinate data on
pipes are retrieved from the GIS. Coordinate data may include x-y
coordinates, street addresses, latitude/longitude coordinates, or
other types of geographical coordinate representations used in GIS
systems. Asset data of pipes and sensors from an asset management
system are retrieved, step 303. The snapping process loops through
each pipe, step 305, and each end of every pipe, step 307, which
are analyzed using the data retrieved from steps 301 and 303. Every
pipe may include a given location, region, DMZ, FMZ, or any
specified network of pipes. Steps 305 and 307 may be carried out as
a nested for-loop or any other suitable type of recursive
programming statements.
For each pipe end, a determination is made whether the given end of
the pipe is a capped end or a service supply connection, or
otherwise not expected to connect onwards to another pipe or asset,
step 309. A capped end is typically an end of a pipe in which a cap
is placed on the end of the pipe that prevents water from flowing
past beyond the capped end. This information would typically be
stored in the GIS or asset management data, or readily deducible
from it. If the end has a cap, the method proceeds to determine
whether both ends have been processed, step 319. Otherwise,
coordinates of ends of other pipes closest to coordinates of the
pipe end are located from the GIS data, step 311. In step 313, a
determination is made whether the coordinates of the pipe and the
located coordinates closest to the pipe match exactly or nearly
exactly. If the coordinates are exact or matches, the matching
coordinates are stored as pipe junctions, step 317. However, if
they do not match, one or both of the pipe end coordinates are
adjusted to match based on some criteria, step 315, before being
stored as pipe junctions in step 317. Criteria to adjust the
coordinates may include, but are not limited to, relative location,
asset age and material, pipe diameter, etc. In an exemplary
criterion, a given pipe end may be much closer than other pipe ends
in consideration, where the other pipe ends can be snapped to
additional pipes, then the given pipe may pass such a criterion. In
a further example, a given pipe end which is not the closest pipe
end may be the only candidate of matching diameter, and may be
selected based on this criterion. Step 317 may also include raising
an exception for the pipe end coordinates if no match passes the
criteria. For example, if several pipe ends are reasonable
candidates to snap to a given pipe end, the system chooses not to
snap anything, but rather flags the data point for user
attention.
A determination is made whether both ends have been processed, step
319. If not, the method returns to step 307 to process the other
end of the given pipe. When both ends have been processed, a
determination is made whether all pipes have been processed, step
321. If all pipes have not been processed, the method returns to
step 305 to continue processing other unprocessed pipes in the same
fashion. At the completion of this snapping phase of
pre-processing, the ends of the pipes in the network have, wherever
possible, been "snapped" together in the model based on data from
the GIS and asset management system, so that a cohesive and
comprehensive model of the connections between pipes can be
generated. The connections are stored in the database as junctions,
pending further pre-processing described below with reference to
FIG. 4.
Upon processing all the pipes, processing is performed for every
sensor, step 323, in a second, similar phase of pre-processing
involving proper location of sensors in the network. In step 325, a
determination is made whether a given sensor coordinate as
retrieved from the GIS matches a pipe coordinate. If the
coordinates match, the method bypasses steps 327 and 329 and the
sensor data is stored, step 331. Otherwise, a pipe closest to the
sensor is located, step 327. One of skill in the art will recognise
that certain types of sensors may be located along pipes, and other
types may be located only at pipe ends (joints) or at other
particular assets or parts of assets; processing proceeds according
to such location requirements for each sensor type. The coordinate
of the sensor is adjusted to match the closest pipe based on
criteria, step 329. Following steps 327 and 329, the connection
between sensor and pipe is stored in step 331. Criteria for sensor
connection may be similar to the ones described for pipe ends.
Storing a sensor connection may include storing a sensor
identification, coordinate, or adjusted coordinate in the GIS or
asset management record of the pipe, and vice versa. In step 333, a
determination is made whether all sensors have been processed. If
not, the method returns to step 323 to continue processing the
remaining sensors not yet processed. When all the sensors have been
processed, the sensors positioned throughout the network have been
properly located within the pipe or joint in which they are
installed. Although this processing has been described for the
improvement of GIS data on sensor locations and their connection to
pipes, one skilled in the art will appreciate that the same or
similar method may be employed to find the logical or connection
relationship between any two types of assets, as required for
analysis of the network layout.
In a next phase of pre-processing of GIS and asset management data,
the pipes and sensors preprocessed according to the methods of FIG.
3 are organized into higher level structures such as DMAs. As shown
in FIG. 4, this processing is performed for every pipe junction
found through snapping of pipe ends during pre-processing, step
401. Every pipe junction entry may include a given location,
region, DMZ, FMZ, or any specified network of pipes and assets. For
each junction, a determination is made whether there the junction
is between only two other junctions, step 403. For example, if a
first junction is found to be connected to a second junction via
two pipes connected only by a third junction, then the connection
between the first and second junctions is merged into a single
logical pipe, step 405, and the third junction is eliminated as a
junction. The merged connection is stored in a database, step 407.
If the given junction is not between only two other junctions, then
the method proceeds to determine whether there are any other
junctions to process, step 409. Upon determining that there are
still junctions to be processed, the method returns to step 401 to
process the remaining unprocessed junctions in the same fashion.
When all junctions have been processed, pipe connections and
junctions are collected into DMAs or FMZs based on valve locations
and status, step 411. The system traverses the network graph,
starting from the known inlets of the DMA, or from other points
known to be part of the DMA, such as certain listed sensors or
other assets, and treating capped pipes, closed valves, and flow
meters as "disconnects" or boundaries. Any method for finding
"connected graph components" may be used, as known to one skilled
in the art. The resulting subgraph is taken to be the total
collection of assets in the DMA. The DMA or FMZ status and asset
data are stored in a database, step 413. In one embodiment, the
pipe connections, junctions, statuses, and asset data may be stored
in the enriched GIS database. In some embodiments, the system may
further verify that the resulting subgraph does not extend
significantly beyond the polygon stored in GIS as the approximate
DMA geographical boundary, nor cover significantly less than the
total assets within that polygon. One of skill in the art will
appreciate that geometrical inclusion of asset within the DMA
boundaries is not exactly equivalent to being a logical, functional
part of that DMA, depending on actual hydraulic connectivity.
Following the pre-processing of the GIS and asset data, the
enriched GIS database contains an accurate set of structural and
geographic data from which mathematical graphs and network traces
can be made to view a full topology of the network. The network
trace may be displayed on a user interface. FIG. 5 illustrates an
exemplary network trace depicted as a mathematical graph including
a map of assets, and/or GIS layers or data corresponding to DMAs of
a water distribution network in a region. DMAs depicted on the
network trace may be selected by a network planner where certain
DMAs may be added to or removed from the network trace for display
and analysis. The network graph may show pipes and other assets
with characteristic information of the assets within one or more
DMAs, e.g., length, physical characteristics, diameter, age, etc.,
displayed along with assets, or hidden within the assets and
displayable upon selection, navigation to, or hovering with a
cursor over the assets. A network graph such as this may include
graphical depictions and locations of pipes and assets in a given
DMA. In one embodiment, the pipes and other assets in the network
graph may be searchable. In another embodiment, the characteristics
may be visibly displayed along or adjacent to each asset.
As shown in FIG. 5, DMA 500 includes inlets 502, 503, 504 and 505,
nodes 500a, 500b, 500c, 500e, 500f, 500g, and segment 500d. DMA 501
includes inlets 505, 506, 507 and 508, nodes 501a, 501b, 501c,
501d, 501e, 501f, 501g, and segment 501e. Inlets 502, 503, 504, and
505 are respectively connected to nodes 500c, 500a, 500e, and 500g
in DMA 500, while inlets 505, 506, 507 and 508 are respectively
connected to nodes 501g, 501b, 501a, and 501c. The inlets may
provide a flow of water into the DMAs or out of the DMAs. Inlets
that provide water out of DMAs may also be referred to as outlets.
Inlet 505 is an inlet that is shared between and connects DMA 500
and DMA 501. The illustrated graph also shows that inlet 508
connected to DMA 501, overlaps across, but is not connected to or
part of, DMA 500. Inlet 508 may be a water main that distributes
water to secondary distribution lines such as in DMA 501.
Each of the illustrated lines/edges and inlets 502-508, segments
500d and 501e may be pipes or mains in the DMAs. Nodes 500a, 500c,
500g, 500e, 501a, 501b, 501c and 501g may be meters and sensors,
whereas nodes 500b, 500f, 501d and 501f may be meters and sensors
or any other type of asset such as valves. As known to those of
skill in the art, meters or sensors monitor and measure
characteristics of the fluid flowing through the inlet pipes. Such
characteristics may include flow rate, pressure, temperature, pH,
salinity, chlorine amount and turbidity. F1, F2, F3, F4, F5, F6,
and F7 in FIG. 5 indicate flow rates measured at the corresponding
inlets. F1-F7 may also represent a pressure value or any other
characteristics measured across the pipes.
In some embodiments, the graph or connectivity structure may be
stored as adjacency matrices, adjacency lists (represented by
arrays, linked lists, or hash tables), or any other suitable data
structure known by one of ordinary skill in the art. In another
embodiment of the present invention, mathematical transformation of
the map and/or GIS layers and data may include determining flow
monitoring zone or FMZ consumption equations or formulas. A FMZ is
generally an area or district of a distribution system which is
specifically defined, e.g. by the closure of valves, and in which
the quantities of water or other resources entering and leaving the
district are metered. A FMZ may be described as a sub-network
connected to the rest of the network only through valves which are
routinely closed or through flow meters; DMAs are typically FMZs
deliberately laid out by network planners and used by network
operators for monitoring purposes. As an example, referring to FIG.
5, a maintenance closure of segment 500d in DMA 500 would result in
the creation of two FMZs within DMA 500.
One way to monitor a water system may include tracking consumption
equations, or the sums of flow meters that define FMZs. These
include flow meters at the inlets of the FMZ. Such equations may be
used, for example, to monitor for leaks. Consumption formulas
enrich the geographic information about assets for utility
companies. Many water utilities have a partial or inaccurate list
of consumption equations. Identifying the flow meters along the
boundaries of a FMZ is a product of performing a proper network
trace, but is insufficient, as flow meters either measure flow in
one direction only, or have a direction measured as positive flow,
and one measured as negative. These directions are rarely recorded
reliably and may change occasionally during maintenance as meters
may be rotated and replaced without proper recording of the
orientation or direction of the meters.
Data collected from meters and sensors at these various locations
is fed into the water network analysis engine as described above.
By virtue of the processes performed on the GIS and asset data by
the structural analysis system of the present invention as
described herein, the network analysis engine now has access to
more accurate structural and topology information on the network
from which to perform its anomaly detection. As a result, the
network analysis engine can produce significantly more accurate
results in the way of improved anomaly detection, better
classification of events, more accurate location of events within
the network, and fewer false positives, among other things.
For example, using GIS data, statistical algorithms, and recent
meter data, water utilities may find additional consumption
equations or correct existing ones. GIS layers or data from a map
are analyzed to find FMZs, which may include sub-networks in which
all the inlets and outlets are metered. Referring to FIG. 5,
consumption of FMZs may be derived and modeled from the enriched
GIS data or from a graph of the network. The directional signs of
each flow meter may need to be determined for each consumption
equation (i.e., should the meter be added or subtracted).
Directions for each flow meter may be represented by the polarity
of the values of F1-F7 of FIG. 5. Generally, a positive flow of
water into a given DMA is represented by a positive value, while a
negative flow of water out of the same DMA is represented by a
negative value. For example, inlet 505 may have a positive `+` sign
F4 or consumption, measured at node 500g, while the opposite end of
inlet 505 may have a negative `-` sign F4 or consumption, measured
at node 501g. Similarly, nodes 500a, 500c, 500e and 500g may have
`+` signs and the sum of positive flow at nodes 500a, 500c, 500e
and 500g should be equal to the sum of negative flow values seen on
the opposite ends of inlets 502, 503, 504 and 505, so that the net
sum of flows on both ends of inlets 502, 503, 504 and 505 is equal
to zero +/- to maintain the law of conservation of mass.
In reality, information concerning flows may be missing and
difficult to determine. However, according to an embodiment of the
present invention, a "best guess" approach may be taken. In
determining the actual consumption flows or signs, probabilities of
possible flow combinations are calculated to match several
criteria, factors, and conditions. For example, to determine a
plausible consumption formula for DMA 500, a plurality of flow
direction combinations is determined by a sum of (+/-) F1, (+/-)
F2, (+/-) F3, and (+/-) F4. The probability of a most correct flow
direction combination may be based on a comparison of actual flow
values from the DMA's flow meters with general DMA day/night
consumption expectations (possibly based on number and
characteristics of service connections), periodicity, historical or
actual consumption data, and variance. When added correctly, a
graph of consumption should result in a smooth curve with the
above-listed characteristics and some additional short-term
variances (random noise). In essence, each of the flow directions
corresponding to the flow values of F1-F4 may be "reverse
engineered" to match an expected net consumption. In one
embodiment, small inlets affecting less than, for example, 1%, may
not matter and can be ignored in determining the consumption
formulas. It is to be noted that while this method may not produce
the exact signs for each flow value, a correct or nearly correct
net result of the consumption formula will be determined.
If flow meter signs cannot be found directly from the GIS data (or
are suspect because of an unreasonable pattern for the sum),
statistical algorithms may be applied to automatically infer signs
from the GIS and sensor data. Unreasonable patterns may be
determined by conducting periodic analysis of meter data (e.g.,
every month) retrieved from the sensor archive database utilizing a
monitoring system and comparing the analysis with the sum.
According to one embodiment for determining flow directions of the
present invention, a combination of all possible signs of meters
may be searched for each FMZ and verified with meter configurations
to produce consistent consumption patterns. Various attributes of
the consumption values may be verified in order to check whether
the consumption values match the typical expected patterns (e.g.,
they should exhibit daily, weekly, seasonal patterns, and the ratio
between the minimum and maximal daily values should be within a
certain range and have a low variance, etc.). Consumption values
may also be verified by determining the statistical likelihood of
the values in light of historical patterns of an examined FMZ (e.g.
showing that "flipping" a single meter direction completely
explains a sudden change in flow patterns), in light of similar
FMZs, or in light of typical FMZ patterns. If more than one
satisfactory configuration exists for some of the FMZs,
configurations may be ruled out by comparing FMZs that share one or
more flow meters. For example, if a flow meter is determined to
have a `+` sign in a FMZ (because the `-` direction doesn't produce
the expected consumption pattern), then the flow meter on the
opposite side must have a `-` sign in the FMZ.
According to another embodiment of the invention, a score may be
generated for each configuration of meter directions in the entire
network to describe how well the resulting consumption values match
the expected patterns, and the best-scoring configuration of meter
directions may be selected. This may be required, in the event that
conflicting "best guesses" for many DMAs--considered
individually--cannot be easily resolved. An exhaustive search for
all possible directions for network meters may not be feasible,
unlike a search for five meters in a single FMZ, for example.
Accordingly, optimization procedures or algorithms known by one of
ordinary skill in the art may be applied to determine the possible
sign combinations. This could be by means of a genetic algorithm or
simulated annealing in order to find an optimal solution. In
another embodiment, combinatorial optimization may be used to find
an optimal object from a finite set of objects, in this case, flow
directions.
FIG. 6 presents a network graph for characterizing a water
sub-network according to an embodiment of the present invention.
FIG. 6 illustrates a mathematical network graph of mathematical
graph elements based on enriched GIS and asset data including FMZ
600 and inlets 602, 603, and 604. A proposed inlet 605 is
illustrated by a dash edge or pipe and does not exist as an actual
asset in FMZ 600. Further details of proposed inlet 605 will be
described in further detail below with respect to the discussion of
single point of failures. FMZ 600 further includes nodes 600a,
600b, 600c, 600e, 600g, and segment 600d. Each edge may represent a
pipe or main in the FMZ and each node may present another type of
asset such as meters. Inlets 602, 603, and 604 are respectively
connected to nodes 600a, 600b, and 600g in FMZ 600. The inlets may
provide a flow of water into the FMZ or out of the FMZ.
In some embodiments, the mathematical network graph is analyzed by
a network analysis engine and the results of the analysis are
displayed on a user interface. Analysis of the mathematical graph
may include characterizing FMZs on the mathematical graph.
According to one embodiment of the present invention, "Split FMZs"
may be identified on the mathematical graph, which consists of
several separate sub-networks having parts that are hydraulically
disconnected, or connected via a flow meter. The split FMZs may
also be viewed as smaller FMZs within a FMZ. For example, two
sub-FMZs may be created when the network connected to segment 600d
at 600c and the network connected to segment 600d at 600e become
hydraulically disconnected by virtue of the pipe represented by
segment 600d being closed. Sub-FMZs typically occur when, though
network planners intended to create a single DMA, subsequent
maintenance, faults, and shorter-term planning break the DMA at
additional points or lead to permanently closing additional
valves.
Recognizing split FMZs is useful in monitoring a water system. For
example, a net consumption of water in a water utility network may
be computed from a subset of an original FMZ's meters. An exact
equation for calculating consumption for each sub-FMZ may be
determined from or inferred from the mathematical graph, as
described in further detail below. Thus, each sub-FMZ may be
monitored separately using analytical techniques disclosed from the
previously mentioned U.S. Pat. No. 7,920,983. This may improve the
sensitivity of the anomaly detection algorithms since smaller
overall consumption values may be monitored, making an anomaly of
given magnitude appear to stand out more in comparison. For
example, referring to FIG. 5, instead of monitoring F1, F2, F3, and
F4 in DMA 500, F1, F2 and F3, F4 may be monitored separately if
segment 500d were to be hydraulically disconnected. Geographical
resolution of anomaly detection may also improve directly, since
the sub-FMZs in which detection may occur are smaller.
Similarly to the split FMZs situation described above, another
characterization may include FMZs that are almost split into
smaller FMZs ("almost split FMZs"). As shown by virtue of the
network trace in FIG. 6, the only hydraulic connection between the
rest of FMZ 600 on the other end of segment 600d and the two
separate sub-networks is through a pipe from node 600c to 600e.
Flow into the two separate sub-networks may be bound by or limited
to the flow of the pipe at node 600c. FIG. 6 illustrates FMZ 600
containing a relatively long and narrow pipe (segment 600d) that
parts into two separate sub-networks at node 600c and at 600e. The
length and diameter of the pipe would have been retrieved from the
asset management data and is available to the water network
analysis engine from the enriched GIS database. As one skilled in
the art will appreciate, given the typical flows and pressures of
reasonable operation ranges in and around the FMZ, these parameters
define a rough range within which the flow through that pipe may
fall. In such a case, if the system determines that the flow in
segment 600d is necessarily below a predefined threshold or
proportional threshold, for example representing a small flow
compared to the total flow into the FMZ, the two sub-networks may
be treated as two "almost separate" FMZs. Each of the separate
sub-networks or newly created FMZs may be monitored separately,
with some known level of inaccuracy. For example, if an abnormal
change is detected in the consumption of a given one of the
sub-networks/almost split FMZs, and the change is larger than the
maximal flow in the pipe segment 600d connecting the sub-networks,
then the anomaly may be due to an event in the given one of the
sub-networks since the change cannot be attributed to the water
flowing from or to the segment pipe 600d.
According to another embodiment of the present invention,
characterization of FMZs on a network graph such as the one
illustrated in FIG. 6 may also reveal sub-networks having single
feeds, which may be referred to as a "single point of failure."
Such situations may be problematic, since an entire area could
potentially be left with insufficient water supply or very low
pressure if the single pipe feed were to be broken. This situation
may result in the failure to meet a required service requirement
and is costly to a water utility. Single point of failures are not
limited to areas with single feeds to sub-networks, but may also
apply to areas where there are inadequate amounts of pipes or valve
openings feeding a plurality of sub-networks. For example, if inlet
604 were to be shut down or closed, a significant portion of FMZ
600 on the right side of segment 600d may be dramatically
affected.
According to one embodiment of the present invention, a water
utility planner may desire to make changes to a network to correct
these issues and avoid single points of failure, such as by
installing additional pipes, opening network valves, replacing
pipes with ones having thicker diameters, or routing other pipes
and valves to a single-feed region. Using the enriched GIS data
generated by virtue of the present invention, recommendations for
correcting single points of failure may be generated by the water
network analysis engine. Single point of failure analysis and
recommendations may include determining whether there is reasonable
redundancy in the network in case of a point failure caused by a
burst or closing of valves for any reason. The water analysis
engine would perform graph analysis on the mathematical network
graph generated by the structural analysis engine to determine and
recognize single point of failure locations. The single point of
failure locations may be displayed or reported to a user monitoring
the water network and recommendations may be provided for resolving
these issues.
For example, the recommendation may include providing a proposal
for adding inlet 605. Inlet 605 may be shown on the user interface
highlighted, emphasized, or, as in the example of FIG. 6, dashed to
indicate to the user a recommended location for adding a pipe or
feed to a sub-network of pipes. An indication of a maintenance
recommendation including inlet 605 may be displayed on a user
interface. The recommendation may be displayed on the user
interface as a network trace of the sub-network, as illustrated in
FIG. 6, or on a map with one or more GIS layers. In one embodiment,
a user viewing the user interface may select to view a water
utility network in either network trace or map form and may switch
from one form to the other. Delivery of recommendations to a user
may be displayed as a graph or list and stored in a database such
as the enriched GIS database. In an alternative embodiment,
recommendations may be delivered to a user of the water network
analysis engine periodically based on a schedule defined either by
the user or the water network analysis engine.
Similar to an "almost split FMZ," there may also be an "almost
single point of failures," where alternative feeds, if a single
point of failure were to fail, are incapable of supplying the
area's consumption under a typical hydraulic scenario. Almost
single point of failures may be subjected to the same
recommendations as single point of failures. As discussed above, if
inlet 604 were to be shut down or closed. A significant portion of
FMZ 600 on the right side of segment 600d may be dramatically
affected. According to one embodiment, segment 600d may be
indicated as an almost single point of failure if inlet 604 were to
be closed, as segment 600d, carrying any flow from inlets 602
and/or 603, may be unable to supply the demand previously supplied
by inlet 604.
FMZs may also be characterized as neighbors if there is at least
one (closed, unmetered) valve connecting between them. If one of
these valves is opened, intentionally or by accident, the two FMZs
become hydraulically connected, and since there is no flow meter at
the connection, the FMZ consumption formulas are no longer correct.
As an example illustrating neighboring FMZs, the two sub-networks
connected by segment 600d at nodes 600c and 600e may be two
neighboring FMZs, where in this example, segment 600d is closed. It
is to be noted that two FMZs may be geographically adjacent, but
they may not necessarily be hydraulic neighbors (neighboring FMZs),
while two FMZs may be geographically distant, yet they may be
hydraulic neighbors. According to one embodiment of the present
invention, analysis of a mathematical network graphs may include
analyzing all the FMZs on the mathematical graph to automatically
generate a table of neighboring FMZs. This table may be utilized
for identifying and classifying anomalies in the network. For
example, a breach between two FMZs may be reported if an abnormal
flow increase is observed in a FMZ and a decrease of similar
magnitude is observed in a neighboring FMZ, where both anomalies
appear to have begun simultaneously or almost simultaneously. DMA
or pressure-zone breaches are a common and costly fault in water
utility network operation. In one embodiment, the characterizations
of FMZs (e.g., split FMZs, neighboring FMZs, etc.) may be used in
the analysis of consumption in determining the consumption
equations described above.
In one embodiment, analysis of a mathematical network graph may
include choosing sensor locations to enable automated monitoring or
to improve existing automated monitoring. A water network analysis
system may help decide for a network planner how many sensors to
add, in which DMAs or areas, and in which precise locations, in a
manner most helpful for future monitoring, especially automatic
monitoring of utility networks. As explained in the previously
mentioned U.S. Pat. No. 7,920,983 and commonly owned U.S. patent
application Ser. No. 13/008,819, network meters, such as flow and
pressure sensors, allow the water network analysis engine to not
only compute water consumption balances and perform various network
operation tasks, but also to identify anomalies such as leaks,
breaches, etc., and estimate their locations. Using analysis of GIS
layers of a FMZ and data collected from existing network meters,
automated recommendations may be made for locations to install new
meters. The locations may be selected in a way that maximizes the
benefits gained from the new meters with respect to the detection,
classification, and geolocation of anomalies.
The methods described herein are similarly useful in determining
optimal meter locations for installing additional temporary or
permanent acoustic sensors in locations within the network for
improving the accuracy and reliability of hydraulic models derived
from the GIS data. These sensors may be temporary or mobile
sensors, used to collect data for a short period, for example a few
days, at a particular location, then moved to the next region to be
examined. This provides for additional monitoring for periodic
recalibration of the hydraulic models for a network. Data gathered
from the sensors may indicate inaccuracies of the hydraulic models
(when sensor data does not match model predictions) and help
correct them. Results from hydraulic models should bear close
resemblance to the actual performance of a hydraulic system. In
other words, modelling of a network, such as a water utility
network, should be calibrated to provide results bearing a close
resemblance to reality. Without such frequent calibration, the
model may be of limited value.
Exemplary methods for determining optimal locations for meters are
now described in further detail. An FMZ may be partitioned into
several smaller sub-areas, so that the flow into each of these
sub-areas can be computed or estimated. In one embodiment, a
partition may be made in which the sub-areas are more or less
similar in size (i.e., balanced cuts) to achieve the best possible
monitoring performance (e.g., to be able to accurately detect small
leaks) in each of the sub-areas. The size of an area can be
measured using: water consumption, in order to optimize anomaly
detection in the meters data; total length of pipes, to minimize
the cost of leak detection by acoustic teams; cost of step-testing;
and a combination of the above, for example a weighted linear
sum.
Considering a graph G(N, E), where N denotes the set of nodes and E
the set of edges, the balanced cut problem may be viewed as: given
G and an integer k>1, partition N into k parts (subsets) N1, N2,
. . . , N.sub.k such that the parts are disjoint and have equal
size, and the number of edges with endpoints in different parts is
minimized. A cut of size k in a connected graph is a set of k edges
whose removal would partition the graph into several separate
connected components. A cut of size k may be selected such that the
size of the largest component is minimized. In other words, a k
amount of pipes in a mathematical graph of the FMZ is selected, so
that if the flow in these pipes is metered, the system may
effectively obtain several separate sub-areas whose consumption is
monitored, and where the sub-areas are as balanced as possible,
i.e., the size of the largest sub-area is minimal.
Sub-areas may be required to be smaller than a given cutoff
threshold, and should be achieved with a small k value. "Balanced
cut" algorithms known by one of ordinary skill in the art may be
used to find optimal balanced cuts using the above measures of
size. If the FMZ and the parameter k are relatively small, a
brute-force exhaustive search (of edges) may be used. Otherwise,
standard heuristics or optimization techniques (e.g. a genetic
algorithm) may be used to converge to a good solution. In one
embodiment, a water utility may select to install f new flow
meters, then k shall be set to equal f. One flow meter may be
installed in each pipe of an optimal (balanced) cut.
For example, referring to FIG. 6, a cut of size k=1 may be
selected, where k is equal to the number of additional flow meters
to install, and the edge between node 600b and 600c may be
identified as the edge to "cut." This cut carves out a smaller FMZ
to provide better monitoring and detect smaller leaks that may not
be possible in a larger FMZ. A water analysis engine may indicate
the installation of a meter at the edge between node 600b and 600c
as an optimal location. An additional constraint may be a list of
pipes (or other particular assets or locations), in which it is
possible to install a flow meter. In this case, the search for a
balanced cut may be limited to the listed possible locations.
According to another embodiment of the present invention, in the
case p pressure loggers to be installed, k may be set to equal p/2.
That is, for each of the k pipes in the balanced cut, the water
utility should install two pressure loggers--one on each end of the
pipe. From the hydraulic parameters of a pipe (diameter, length,
and roughness), it is possible to estimate the flow along the pipe
by comparing the pressure decrease between the two loggers and
applying standard hydraulic equations (e.g., Hazen-Williams or
Darcy-Weisbach). Two pressure loggers on each pipe essentially act
as a flow meter for monitoring the flow between the sub-areas
associated with each pipe. For example, pressure loggers may be
installed along the edges above node 600e to serve as a flow meter
that will monitor flow to proposed inlet 605. Pressure sensors may
be significantly cheaper and easier to install than flow sensors,
making this a desirable scenario.
Some pipes may be inadequate for this type of analysis, since the
flow cannot be computed within reasonable accuracy based on the
pressure drop along the pipe (e.g., if the pipe is very short or
wide). In such a case, these pipes may be excluded from the k-cut
search. In a common scenario, pressure loggers may only be
installed in a pre-defined set of hydrants. A pipe in the network
will be included in the analysis only if there are hydrants on, or
hydraulically very close to, its ends. Other constraints on
possible locations for installing pressure loggers may be handled
similarly.
Optimal locations for installing f flow meters and p pressure
loggers can be found using a combination of the above methods. The
analysis may be aimed at network planning which serves to improve
monitoring. One embodiment includes the selection of optimal
locations for additional sensors in a DMA to enable better
automatic monitoring through, for example, a system as disclosed in
the previously mentioned U.S. Pat. No. 7,920,983.
According to another embodiment of the present invention, pressure
loggers may be installed in such a way that will provide a
reasonable approximation of pressure values at every point in the
FMZ(s). These pressure values may then be compared to their
standard distribution (e.g., using previous days or weeks) of
pressure values. An abnormal decrease in pressure in a certain area
within the FMZ implies increased flow to that area, e.g., due to a
leak. More accurate and robust results can be attained with a
hydraulic model that is calibrated using the data from all the
meters.
In order to place p pressure loggers, a k-center problem on the
network trace of the FMZ may be solved, with k=p (pressure
loggers). That is, a k-center is a set of k "center" nodes in the
graph, such that the largest distance from any node in the network
to its nearest center node is minimal. A pressure logger may be
installed at each center of a k-center solution. The distance
measure used may be an approximation of the hydraulic distance
between two points, e.g., the expected pressure drop between the
two points according to the Hazen-Williams formula, assuming some
fixed flow F between the points. In an alternative embodiment, the
average flow between two points under normal conditions may be used
as the flow F between the points, in case this flow can be
estimated, e.g., from a hydraulic simulation. Heuristics and
approximation algorithms known by one of ordinary skill in the art
may be used for solving the k-center problem. If the pressure
loggers can be installed only in a pre-defined set of points (e.g.,
hydrants), then a variant of the k-center problem may be solved,
wherein the centers can only be placed in a subset of the graph's
nodes.
In order to detect hidden ("background") leaks or prevent future
leaks, utilities may carry out large-scale acoustic surveys or
maintenance projects, such as replacing pipes and other equipment.
Both tasks are labor intensive and costly, so identifying the best
target locations is extremely important. Leakage-prone areas within
the network that should be targeted for maintenance or acoustic
inspection may be identified and prioritized by a combined analysis
of the GIS layers using the enriched GIS data and the online
metered data. Below are examples of features that might increase
the rate of leaks in a specific area (e.g., FMZ or pressure managed
zone): average (and/or maximum) age of hardware, mainly pipes,
fittings and valves; network complexity, e.g., number of pipes,
fittings, valves (of various types), service connections;
geographical properties that contribute to more leaks being hidden
(as opposed to visible), e.g., areas near rivers, coasts, or areas
with soft land; high pressure, or large variability of pressure, as
inferred from: a. topography--large differences in elevation within
the same pressure zone imply that the pressure in the low elevation
points is probably high; b. pressure meters readings; high rate of
leaks or bursts, estimated from: a. historical repairs files; b.
computational analysis of historical data from flow and pressure
meters; several leak-related parameters may also be analyzed, such
as: leak frequency--total number of leaks per day per ft of pipe;
water loss rate--total amount of water lost per day per ft of pipe;
hidden time--average time it took for a leak to be detected
(estimate the leak's start time by analyzing the historical meter
data); and total cost--sum of all relevant costs, e.g., due to
water loss, repairs, collateral damage, regulator fines.
Another way in which the enriched GIS data generated by virtue of
the present invention may be used to improve operation of a water
utility network is for prioritizing areas for maintenance. FIG. 7
presents a flow diagram of just such a method for prioritizing
according to an embodiment of the present invention. In order to
choose the best parameters for the algorithm and combine the
information from the various features, machine-learning may be
used. In step 701 a training set of historical leak data for a
first time period of the utility network is received. The training
set may include recorded locations of previous leaks, bursts, or
failures, and an associated leak rate for a duration of the leaks
at the recorded locations. Next, in step 703, one or more feature
values are retrieved from enriched GIS database for a time period
corresponding to the first time period. The features may include
the features described above that may increase or influence the
rate of leaks in a specific area. For example, a rate (or cost) of
leaks in the selected network areas may be estimated using
historical repair records and/or leak detection methods. Extraction
and computation of the features are preferably taken for a time
period before the period of the training set. The features may be
obtained from enriched GIS data and/or an analysis of a
mathematical graph created from the enriched GIS data, created as
described above.
In step 705, the utility network is partitioned into regions with
fairly uniform characteristics. The characteristics may include,
for example, pipe age, length, soil type, number of junctions, etc.
A model is generated from the training set, the one or more feature
values, and characteristics, step 707. The model may be generated
by running a machine-learning black-box to optimize the model's
parameters, and to find a model that accurately predicts the
leakage rate from the feature values associated with the training
set. Various machine-learning approaches may be used to choose a
best performing model, such as linear or non-linear regression,
decision trees, neural networks, k-nearest neighbor, support vector
machines, and other optimization techniques.
Once a model is generated, the model is provided with one or more
current or new associated feature values and characteristics, step
709. In a next step 711, predicted leakage or failure rates are
obtained for one or more regions from the model. An index may be
generated storing network areas and the associated predicted
leakage rates. The results of the predicted leakage rates are
stored and reported in step 713. The predicted leakage rates may be
used by utility companies to prioritize field work or preventive
maintenance, possibly after taking into account additional criteria
such as the varying cost of work in different areas. Areas or
locations with higher leakage or failure rates may be assigned
higher priorities for repair.
Another advantageous use of the enriched GIS data generated by the
structural analysis engine of the present invention is in
developing and implementing step-testing. Step-testing is an
analysis technique used for leak detection that involves shutting
down or closing off specific parts of a network in sequence to make
artificially defined zones and monitoring water flow in and around
those parts or zones until a leak is detected or isolated within
the specific zone. It is especially useful when acoustic techniques
are inapplicable or ineffective, e.g., due to background noise or
pipe materials that absorb the leak's noise. More specifically,
step-testing may include several steps or phases where, in each
step, a different part of a targeted sub-network such as a FMZ is
shut down by stopping water supply to that area for a certain
length of time by closing one or more valves. Water flow into the
FMZ may be monitored during the test and then analyzed. For each
step, an expected decrease of flow under normal operating
conditions should be proportional to the size of the area that was
shut down or, more accurately, the number and size of service
connections in the area. If the area includes a leaking pipe, an
additional decrease is expected, equal to the leak's magnitude. If
an abnormally high decrease is observed in several steps, the leak
can be pinpointed to the overlap of the respective areas of the
steps. Step-testing is usually carried out at night, when water
consumption is low and stable, random variations are minimal, and
disruption to consumers is relatively light.
Network traces may also be used to plan for optimal network
monitoring operations, such as a step-testing plan. The enriched
GIS data or network traces can be used to enhance the calculation
of step-testing costs and help determine better locations for
future assets. The improvements in step-testing may include, for
example, the reduction of valves that need to be open and closed,
and a reduction of steps taken. Enriched GIS data can help
determine optimal ways to deploy new assets such as meters,
sensors, valves, and pipes in a water utility or other type of
network to enable optimal future step-testing, when leaks
appear.
FIGS. 1 through 7 are conceptual illustrations allowing for an
explanation of the present invention. It should be understood that
various aspects of the embodiments of the present invention could
be implemented in hardware, firmware, software, or combinations
thereof. In such embodiments, the various components and/or steps
would be implemented in hardware, firmware, and/or software to
perform the functions of the present invention. That is, the same
piece of hardware, firmware, or module of software could perform
one or more of the illustrated blocks (e.g., components or
steps).
It should also be understood that the invention applies not only to
water utility networks, but to any type of distribution system.
Other types of distribution systems may be: oil, wastewater or
sewage, gas, electric, telephony, heating ventilating and air
conditioning (HVAC) systems, or other energy delivery systems which
involve fluid or flowing resources from one area to consumers.
Indeed, the invention may be applied to any distribution or
collection system having meters or sensors at arbitrary locations
in the network measuring distribution parameters such as flow,
pressure, quality or the flow of data itself.
In software implementations, computer software (e.g., programs or
other instructions) and/or data is stored on a machine readable
medium as part of a computer program product, and is loaded into a
computer system or other device or machine via a removable storage
drive, hard drive, or communications interface. Computer programs
(also called computer control logic or computer readable program
code) are stored in a main and/or secondary memory, and executed by
one or more processors (controllers, or the like) to cause the one
or more processors to perform the functions of the invention as
described herein. In this document, the terms "machine readable
medium," "computer program medium" and "computer usable medium" are
used to generally refer to media such as a random access memory
(RAM); a read only memory (ROM); a removable storage unit (e.g., a
magnetic or optical disc, flash memory device, or the like); a hard
disk; or the like.
Notably, the figures and examples above are not meant to limit the
scope of the present invention to a single embodiment, as other
embodiments are possible by way of interchange of some or all of
the described or illustrated elements. Moreover, where certain
elements of the present invention can be partially or fully
implemented using known components, only those portions of such
known components that are necessary for an understanding of the
present invention are described, and detailed descriptions of other
portions of such known components are omitted so as not to obscure
the invention. In the present specification, an embodiment showing
a singular component should not necessarily be limited to other
embodiments including a plurality of the same component, and
vice-versa, unless explicitly stated otherwise herein. Moreover,
applicants do not intend for any term in the specification or
claims to be ascribed an uncommon or special meaning unless
explicitly set forth as such. Further, the present invention
encompasses present and future known equivalents to the known
components referred to herein by way of illustration.
The foregoing description of the specific embodiments will so fully
reveal the general nature of the invention that others can, by
applying knowledge within the skill of the relevant art(s)
(including the contents of the documents cited and incorporated by
reference herein), readily modify and/or adapt for various
applications such specific embodiments, without undue
experimentation, without departing from the general concept of the
present invention. Such adaptations and modifications are therefore
intended to be within the meaning and range of equivalents of the
disclosed embodiments, based on the teaching and guidance presented
herein. It is to be understood that the phraseology or terminology
herein is for the purpose of description and not of limitation,
such that the terminology or phraseology of the present
specification is to be interpreted by the skilled artisan in light
of the teachings and guidance presented herein, in combination with
the knowledge of one skilled in the relevant art(s).
While various embodiments of the present invention have been
described above, it should be understood that they have been
presented by way of example, and not limitation. It would be
apparent to one skilled in the relevant art(s) that various changes
in form and detail could be made therein without departing from the
spirit and scope of the invention. Thus, the present invention
should not be limited by any of the above-described exemplary
embodiments, but should be defined only in accordance with the
following claims and their equivalents.
* * * * *
References